Measuring and mitigating language model biases in abusive language detection

Song, Rui; Giunchiglia, Fausto; Li, Yingji; Shi, Lida; Xu, Hui

doi:10.1016/j.ipm.2023.103277

Cited by 7 publications

(2 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While efforts to address these issues are ongoing, physicians should be aware that human bias is an integral component of any LLM and should be accounted for rather than ignored. 85 On the other hand, as more and more data available online are produced by software, from simpler automated bots to LLMs, it is also true that this will also represent a novel source of bias, with the risk of harming the training of future models by further diluting the quality of available data and reducing the models’ ability to meet human needs and expectations. 86…”

Section: Ethical and Regulatory Considerationsmentioning

confidence: 99%

Large language models in radiology: fundamentals, applications, ethical considerations, risks, and future directions

Akinci D’Antonoli,

Stanzione,

Bluethgen

et al. 2024

dir

View full text Add to dashboard Cite

With the advent of large language models (LLMs), the artificial intelligence revolution in medicine and radiology is now more tangible than ever. Every day, an increasingly large number of articles are published that utilize LLMs in radiology. To adopt and safely implement this new technology in the field, radiologists should be familiar with its key concepts, understand at least the technical basics, and be aware of the potential risks and ethical considerations that come with it. In this review article, the authors provide an overview of the LLMs that might be relevant to the radiology community and include a brief discussion of their short history, technical basics, ChatGPT, prompt engineering, potential applications in medicine and radiology, advantages, disadvantages and risks, ethical and regulatory considerations, and future directions.

show abstract

Section: Ethical and Regulatory Considerationsmentioning

confidence: 99%

Large language models in radiology: fundamentals, applications, ethical considerations, risks, and future directions

Akinci D’Antonoli,

Stanzione,

Bluethgen

et al. 2024

dir

View full text Add to dashboard Cite

show abstract

“…Song. [9] In this research, a variety of measures are employed to assess the existence of bias in machine translation and examine the in uence of these inherent biases on automatic abusive language identi cation. Based on this statistical analysis, two distinct debiasing procedures are suggested, token debiasing as well as sentence debiasing, that are applied in tandem to minimize the bias of machine translation in abusive words detection while maintaining classi cation accuracy.…”

Section: Related Workmentioning

confidence: 99%

Enhancing Communication and Comprehension for Individuals with Special Needs through Federated Learning: A Deep Learning Approach

Elsayed,

Elrashidy,

EL-Sayed

et al. 2023

Preprint

View full text Add to dashboard Cite

Individuals with special needs most of the time find it harder to identify hazards and dangers as well as circumstances that are socially challenging. Hence, they face the risk of falling victim to abuse and violence. In this paper, the main goal is to help people with special needs to more successfully communicate with others and comprehend their surroundings. Machine learning-based solutions are used to help people with special needs in their communication tasks. The proposed machine learning model contains a convolutional layer, attention layer, and Bidirectional long short-term memory (BiLSTM) layer and achieves 99.00% accuracy performance. We applied federated learning to preserve privacy and to help researchers overcome problems they face when dealing with people with special needs.

show abstract

Mitigating social biases of pre-trained language models via contrastive self-debiasing with double data augmentation

Li,

Du,

Song

et al. 2024

Artificial Intelligence

View full text Add to dashboard Cite

Measuring and mitigating language model biases in abusive language detection

Cited by 7 publications

References 5 publications

Large language models in radiology: fundamentals, applications, ethical considerations, risks, and future directions

Large language models in radiology: fundamentals, applications, ethical considerations, risks, and future directions

Enhancing Communication and Comprehension for Individuals with Special Needs through Federated Learning: A Deep Learning Approach

Mitigating social biases of pre-trained language models via contrastive self-debiasing with double data augmentation

Contact Info

Product

Resources

About